¹Laboratory of Ultrastructure Research, Aichi Cancer Center Research Institute, 1-1 Kanokoden, Chikusa-Ku, Nagoya 464-8681, Japan

²Pathophysiology Unit, Aichi Cancer Center Research Institute, 1-1 Kanokoden, Chikusa-Ku, Nagoya 464-8681, Japan

³Department of Surgery II, Gifu University School of Medicine, Tsukasamachi 40, Gifu 500-8705, Japan

Correspondence to: Takashi Takahashi, Laboratory of Ultrastructure Research, Aichi Cancer Center Research Institute, 1-1 Kanokoden, Chikusa-ku, Nagoya 464-8681, Japan

The mitotic checkpoint is thought to be essential for ensuring accurate chromosome segregation by implementing mitotic delay in response to a spindle defect. To date, however, very little data has become available on the defects of the mitotic checkpoint in human cancer cells. In the present study, impaired mitotic checkpoint was found in four (44%) of nine human lung cancer cell lines. To our knowledge, this is the first demonstration of frequent impairment of the mitotic checkpoint in this leading cause of cancer deaths. As an initial step towards elucidation of the underlying mechanism, we further undertook a search for mutations in a key component of the mitotic checkpoint, known as hsMAD2, and its immediate downstream molecule, p55CDC. No such mutations were found, however, in either 21 lung cancer cell lines or 25 primary lung cancer cases, although we could identify silent polymorphisms and the transcribed and processed hsMAD2 pseudogene that was subsequently mapped at 14q21-q23. The present observations appear to warrant further investigations, such as search for alterations in other components, to better understand the molecular pathogenesis of this fatal disease, and warn against potential misinterpretation when performing mutational analyses for other cancer types based on cDNA templates.

mitotic checkpoint; lung cancer; hsMAD2; p55CDC; pseudogene

Introduction

Lung cancer has long been the number one cause of cancer deaths in the United States (Minna et al., 1997), while it is also expected to become the leading cause in Japan as it currently claims more than 40 000 lives annually (Statistics and Information Department, 1998). Molecular analyses have demonstrated that a number of genetic lesions in both dominant oncogenes and tumor suppressor genes play crucial roles in the genesis and progression of lung cancers (Minna et al., 1997). In addition, previous cytogenetic studies have shown that lung cancer cells frequently contain complex chromosomal abnormalities including multiple numerical and structural changes, as are also seen in other epithelial cancers (Testa, 1996).

The mitotic checkpoint is a highly conserved mechanism that regulates the cell division cycle and prevents cells with a perturbed spindle assembly from leaving mitosis, thereby improving the fidelity of chromosome segregation (Rudner and Murray, 1996; Paulovich et al., 1997; Elledge, 1998). This mechanism has been intensely studied, especially in yeast, resulting in the identification of several genes that are involved in the regulatory process to maintain accurate chromosome segregation (Hoyt et al., 1991; Li and Murray, 1991). Notably, it was recently reported that human colon cancer cells with chromosomal instability phenotype carried defects in their mitotic checkpoint, suggesting a possible link between impaired mitotic checkpoint and oncogenesis (Cahill et al., 1998).

In the present study, we report identification of frequent impairment of the mitotic checkpoint in human lung cancer cell lines and also describe our initial attempt to elucidate the underlying mechanism by searching for alterations in the mitotic checkpoint genes, hsMAD2 and p55CDC.

Results and Discussion

Frequent impairment of the mitotic checkpoint

Mitotic checkpoints were examined in nine lung cancer cell lines with the aid of nocodazole, which inhibits spindle-kinetochore interaction by microtubule disruption. In addition, we included HeLa and HCT116 which have been shown to exhibit an intact mitotic checkpoint function as well as SW480 and T47D which have been reported to carry defects in their mitotic checkpoint (Cahill et al., 1998; Li and Benezra, 1996). Cells were cultured in the presence of nocodazole for 18 h and examined by means of flow cytometric analysis (Figure 1a). All cell lines exhibited nearly complete cell-cycle block with a DNA content of 4N, where 2N represents the DNA content of cells in the G1 phase. Analysis of mitotic indices, however, indicated significant differences among the lung cancer cell lines (Figures 1b and 2). Four (44%) of the nine cell lines, SK-LU-1, PC-1, ACC-LC-319, and PC-10, exhibited a significant reduction of cells arrested at mitosis comparable to the mitotic indices of the control cell lines with known mitotic checkpoint defects (SW480 and T47D), while SK-LU-1 showed an almost complete lack of mitotic arrest. The remaining five lung cancer cell lines responded well to nocodazole treatment, as did the HeLa and HCT116 lines.

To date, very little data has become available on the defects of the mitotic checkpoint in human cancer cells. Cahill et al. (1998) found such defects in six colon cancer cell lines including SW480, while Li and Benezra (1996) reported them for a single cell line of breast cancer (T47D) and of rhabdomyosarcoma. The present study clearly demonstrates for the first time that the mitotic checkpoint function is also impaired in a significant proportion of human lung cancer cell lines. Our findings also appear to be consistent with previous observations in that we and others have demonstrated infrequent occurrence of widespread microsatellite instability in lung cancers (Minna et al., 1997; Gotoh et al. (in press)), while an impaired mitotic checkpoint was previously reported to be present in colon cancer cell lines without widespread microsatellite instability (Cahill et al., 1998).

Search for mutations in the hsMAD2 gene

Identification of frequent impairment of the mitotic checkpoint led us to search for mutations in the genes involved in this process. As an initial step, we examined 21 lung cancer cell lines for the presence of mutations in the hsMAD2 gene, a key component in the sensing and executing mechanism that delays the onset of anaphase in response to a spindle defect (Li and Benezra, 1996; Chen et al., 1996; He et al., 1997). In the reverse transcription-polymerase chain reaction-single strand conformation polymorphism (RT - PCR - SSCP) analysis, several lung cancer cell lines exhibited distinct mobility shifts, which could be classified into two groups, i.e., Group 1: SK-LC-10 and ACC-LC-97 and Group 2: PC-1, QG90 and SBC3 (Figure 3). Subsequent sequence analysis revealed that Group 1 corresponded to a silent polymorphism at codon 143 (CCA to CCG) of the hsMAD2 gene, and Group 2 to numerous nucleotide substitutions, insertions and a deletion, suggesting that it may represent an hsMAD2 pseudogene (Figure 4a). This possibility was supported by the results of radiation hybrid mapping analysis using both Stanford G3 and Genebridge 4 panels, which consequently placed hsMAD2 and hsMAD2 at chromosomes 4 and 14, respectively. The Genebridge 4 panel showed a linkage of hsMAD2 with WI-6253 (LOD>3, distance=3.25 cR), while the use of the Stanford G3 panel demonstrated that hsMAD2 was tightly linked to GATA51F02 (LOD=1000, distance=0 cR), thus assigning hsMAD2 to 14q21-q23. We also noted that hsMAD2 appeared to be a transcribed and processed pseudogene, since omission of reverse transcriptase eliminated its RT - PCR amplification but RNase-free DNase pretreatment did not, and because a 710 bp genomic fragment of hsMAD2 , which corresponded to the whole open reading frame of hsMAD2 with four exons (Krishnan et al., 1998), could be amplified as a single exon (Figure 4b and data not shown). Accordingly, the present findings indicate that caution should be exerted to avoid potential misinterpretation when performing mutation analyses based on cDNA templates.

Southern and Northern blot analyses of the same panel did not show any gross alterations (data not shown). In addition, 25 primary lung cancer specimens were also examined by means of RT - PCR - SSCP, but yielded no mutations in the hsMAD2 gene. We concluded that hsMAD2 is altered very infrequently, if at all, in lung cancers.

Search for mutations in p55CDC

We next examined p55CDC, a human homologue of Slp1/Cdc20 that binds to hsMAD2 and functions as the target of the mitotic checkpoint (Elledge, 1998; Weinstein et al., 1994). Dominant mutations in Slp1 and Cdc20 have been identified that could overcome the mitotic arrest transduced by Mad2 (Hwang et al., 1998; Kim et al., 1998). Since these mutations in yeast were clustered within a small region, we carried out direct sequencing of the corresponding region of the human homologue, p55CDC, using PS3 and PAS3 PCR primers, which yielded a silent nucleotide substitution at codon 144 (TAT to TAC). No mutations, however, were identified in the same panel of 21 lung cancer cell lines used for the search for hsMAD2 mutations. RT - PCR - SSCP analysis of a 1.5 kb region covering the entire coding region of p55CDC also failed to identify p55CDC mutations (data not shown).

Recent molecular biological studies have clearly indicated that lung cancer is a disease caused by accumulation of multiple genetic defects in dominant oncogenes and tumor suppressor genes, which are involved in various cellular processes such as cell cycle regulation, growth signal transduction, and induction of apoptotic cell death (Minna et al., 1997). Although it remains to be formally proven, accumulating evidence points to the possibility that impairment of the mitotic checkpoint function may be responsible at least in part for the generation of aneuploidy, which is another characteristic feature of lung cancer cells (Cahill et al., 1998; Lengauer et al., 1997). The present study showing frequent occurrence of mitotic checkpoint impairment consequently warrants further study to elucidate the underlying mechanism.

In this regard, the Mad1, 2 and 3 genes and the Bub1, 2, and 3 genes in yeast are known to be important components of the mitotic checkpoint (Hoyt et al., 1991; Li and Murray, 1991). In addition, it has been suggested that the p53 gene, which is the most frequent molecular target among various genetic alterations thus far identified in human lung cancers (Takahashi et al., 1989), may be involved in the mitotic checkpoint (Cross et al., 1995; Gualberto et al., 1998), although conflicting findings have also been reported in the literature (Lanni and Jacks, 1998; Bunz et al., 1998). Our examination of p53 mutations in the panel examined here revealed the lack of apparent association between the presence of p53 missense mutations and impaired mitotic checkpoint (data not shown), while the present study shows that mutations in a key component of the mitotic checkpoint, hsMAD2, and its immediate downstream gene, p55CDC, are infrequent, if they occur at all, in lung cancer cell lines. In this context, it is interesting that alterations in the human BUB1 gene and a related gene, BUBR1, were recently reported in colon cancer cell lines with mitotic checkpoint defects (Cahill et al., 1998).

Taken together, the present observations showing frequent occurrence of mitotic checkpoint impairment suggest that it may represent a characteristic alteration in common human cancers. Although somatic mutations of the BUB1 gene were not found in the four lung cancer cell lines with impaired mitotic checkpoint (unpublished observation), they also warrant further investigations, such as search for alterations in other components including BUB1 and BUBR1 in a larger cohort, to better understand the molecular pathogenesis of lung cancers. It will be also of interest to investigate whether the selective use of certain chemotherapeutic agents, especially those with microtubule disrupting activity such as Taxol, can be realized on the basis of the status of the mitotic checkpoint in individual cases.

Materials and methods

Cell lines and tumor specimens

Twenty-one (nine small cell lung cancers (SCLCs) and 12 non-SCLCs) lung cancer cell lines were analysed in this study. Details of the derivations and culture conditions of cell lines have been described previously (Takahashi et al., 1986; Hibi et al., 1994). Tumor samples, along with uninvolved lung tissue where available, were collected from 25 patients diagnosed histologically as having lung cancers (eight SCLCs, nine adenocarcinomas, six squamous cell carcinomas, one adenosquamous carcinoma and one large cell carcinoma). All tissues were quickly frozen in liquid nitrogen and stored at -80°C until analysed.

Analysis of mitotic checkpoint

Two SCLC (SBC3 and QG90) and seven NSCLC (ACC-LC-176, ACC-LC-319, NCI-H460, QG56, PC-10, PC-1 and SK-LU-1) cell lines were cultured in the presence of nocodazole, harvested at 6 h time intervals up to 36 h. Two-hundred nM of nocodazole was used for all cell lines except for T47D and QG56, which had 500 and 800 nM of nocodazole added, respectively. Cells were then fixed with 4% formaldehyde, followed by staining with 0.1 g/ml of 4', 6-diamidino-2-phenylindole. To measure the mitotic index (percentage of viable cells arrested in mitosis), at least 300 cells were counted for one measurement using fluorescence microscopy. Each measurement was repeated at least twice. Flow cytometric analysis was also conducted to evaluate the cell-cycle profile of cells harvested after 18 h incubation with nocodazole. Cells were stained with propidium iodide and analysed with the aid of FACScan and Cell Fit-DNA software (Beckton Dickinson, Bedford, MA, USA).

PCR - SSCP analysis

PCR amplification using random primed cDNAs was performed with the aid of oligonucleotide primers in the presence of [³²P]dCTP, followed by electrophoretic separation on 6% nondenaturing polyacrylamide gels both in the presence of 5% glycerol at room temperature and in the absence of glycerol at 4°C. The primer pairs used to amplify hsMAD2 cDNA were: M2S1 (sense; 5'-GGAAGCGCGTGCTTTTGTTT) and M2AS1 (antisense; 5'-TCTTTCAGTTGTTCCACCACA); M2S2 (sense; 5'-TTGCTTGTAACTACTGATCTTG) and M2AS2 (antisense; 5'-ATAAATCAGCAGATCAAATGAAC); and M2S3 (sense; 5'-GATCACAGCTACGGTGACAT) and M2AS3 (antisense; 5'-CCTGATTTCAGGAAAACCACA). The PCR amplification was carried out in the presence (M2S3-M2AS3) or absence (M2S1-M2AS1, and M2S2-M2AS2) of 10% glycerol for 30 cycles (94°C for 45 s, 58°C for 30 s, 72°C for 1 min) after the initial denaturation step (94°C for 5 min).

The primer pairs used to amplify p55CDC were: PS1 (sense; 5'-GGCACCAACTGCAAGGAC) and PAS1 (antisense; 5'-TTTGCCAGGAGTTCGGCC); PS2 (sense; 5'-CG-CCAACCGATCCCACAG) and PAS2 (antisense; 5'-CAG-GTTCAAAGCCCAGGC); PS3 (sense; 5'-CCAGACGCCCACCAAGAA) and PAS3 (antisense; 5'-CAGGATACGGTCTGGCAG); PS4 (sense; 5'-CGGAAGACCTGCCGTTAC) and PAS4 (antisense; 5'-GATCCAGGCCACAGAGGA); PS5 (sense; 5'-GAGCAGCCTGGGGAATAT) and PAS5 (antisense; 5'-ACGTGAACCACT-GGACAG); PS6 (sense; 5'-GCTCCCTAAGCTGGAACA) and PAS6 (antisense; 5'-GAGCACTAGGCCACACAT); PS7 (sense; 5'-TGGCCAGTGGTGGTAATG) and PAS7 (antisense; 5'-ATCCACGGCACTCAGACA); PS8 (sense; 5'-CG-CATCTGGAATGTGTGC) and PAS8 (antisense; 5'-GGCTCATGGTCAGACTCA); and PS9 (sense; 5'-CCAAGGTGGCTGAACTCA) and PAS9 (antisense; 5'-ACTGAGGTGATGGGTTGG). The PCR amplification in the presence of 10% glycerol consisted of 35 cycles (94°C for 30 s, 58°C for 20 s, 72°C for 30 s) after the initial denaturation step (94°C for 5 min). The RT - PCR products of lung cancer cell lines showing distinct PCR - SSCP patterns were cloned into the EcoRV site of pBluescript SKII(-) (Stratagene) after polishing, and the resulting plasmid DNAs prepared from pooled clones were sequenced.

Southern and Northern blot analysis

Southern blot analysis using 5 g of genomic DNAs was performed with an hsMAD2 cDNA probe prepared by means of RT - PCR using the M2S1 and M2AS3 primers as described previously (Hibi et al., 1994). Northern blot analysis was performed using 10 g of the total cellular RNAs with the aid of a probe prepared by means of RT - PCR using M2S1 and M2AS2 as described previously (Kondo et al., 1995).

Chromosomal assignment

Chromosomal assignment of the hsMAD2 and the hsMAD2 genes was accomplished by using PCR analysis of genomic DNAs of the Stanford G3 radiation hybrid panel (Research Genetics, Huntsville, AL, USA) as well as those of the Genebridge 4 radiation hybrid panel (Research Genetics). Both hsMAD2 and hsMAD2 were simultaneously amplified by using M2S3 and M2AS3 PCR primers, and the resulting PCR products were distinguished by means of agarose gel electrophoresis based on the differences in size due to the presence or absence of intronic sequences.

RT - PCR analysis

cDNA was synthesized by using 2 g of total RNA of QG90, which expressed hsMAD2 with and without pretreatment with RNase-free DNase (RQ1 DNase, Promega Corp., Madison, WI, USA) to remove contaminating genomic DNA. cDNA was amplified with the aid of the M2S1 and M2AS1 primers and the resulting PCR products were electrophoresed on 8% sequencing gels. PCR amplification using 200 ng of genomic DNA was also carried out to confirm the efficacy of the DNase-treatment.

Acknowledgements

We would like to thank R Ishida for his helpful suggestions and stimulating discussion. This work was supported in part by a Grant-in-Aid for Scientific Research on Priority Areas from the Ministry of Education, Science, Sport and Culture, Japan and a Grant-in-Aid for the Second Term Comprehensive Ten-Year Strategy for Cancer Control.

Bunz F, Dutriaux A, Lengauer C, Waldman T, Zhou S, Brown JP, Sedivy JM, Kinzler KW and Vogelstein B. (1998). Science 282, 1497-1501. Article MEDLINE

Cahill DP, Lengauer C, Yu J, Riggins GJ, Willson JKV, Markowitz SD, Kinzler KW and Vogelstein B. (1998). Nature 392, 300-303. Article MEDLINE

Chen RH, Waters JC, Salmon ED and Murray AW. (1996). Science 274, 242-246. Article MEDLINE

Cross SM, Sanchez CA, Morgan CA, Schimke MK, Ramel S, Idzerda RL, Raskind WH and Reid BJ. (1995). Science 267, 1353-1356. MEDLINE

Elledge SJ. (1998). Science 279, 999-1000. Article MEDLINE

Gotoh K, Yatabe Y, Sugiura T, Takagi K, Ogawa M, Takahashi T, Takahashi T and Mitsudomi T. (1999). Carcinogenesis (in press).

Gualberto A, Aldape K, Kozakiewicz K and Tlsty TD. (1998). Proc. Natl. Acad. Sci. USA 95, 5166-5171. Article MEDLINE

He X, Patterson TE and Sazer S. (1997). Proc. Natl. Acad. Sci. USA 94, 7965-7970. Article MEDLINE

Hibi K, Yamakawa K, Ueda R, Horio Y, Murata Y, Tamari M, Uchida K, Takahashi T, Nakamura Y and Takahashi T. (1994). Oncogene 9, 611-619. MEDLINE

Hoyt MA, Totis L and Roberts BT. (1991). Cell 66, 507-517. MEDLINE

Hwang LH, Lau LF, Smith DL, Mistrot CA, Hardwick KG, Hwang ES, Amon A and Murray AW. (1998). Science 279, 1041-1044. Article MEDLINE

Kim SH, Lin DP, Matsumoto S, Kitazono A and Matsumoto T. (1998). Science 279, 1045-1047. Article MEDLINE

Kondo M, Suzuki H, Ueda R, Osada H, Takagi K, Takahashi T and Takahashi T. (1995). Oncogene 10, 1193-1198. MEDLINE

Krishnan R, Goodman B, Jin DY, Jeang KT, Collins C, Stetten G and Spencer F. (1998). Genomics 49, 475-478. Article MEDLINE

Lanni JS and Jacks T. (1998). Mol. Cell. Biol. 18, 1055-1064. MEDLINE

Lengauer C, Kinzler KW and Vogelstein B. (1997). Nature 386, 623-627. MEDLINE

Li R and Murray AW. (1991). Cell 66, 519-531. MEDLINE

Li Y and Benezra R. (1996). Science 274, 246-248. Article MEDLINE

Minna JD, Sekido Y, Fong KM and Gazdar AF. (1997). In: Cancer Principles & Practice of Oncology, 5th edn, vol. 1. De Vita Jr VT, Hellman S and Rosenberg SA (eds). Lippincott-Raven: Philadelphia., pp. 849-857.

Paulovich AG, Toczyski DP and Hartwell LH. (1997). Cell 88, 315-321. MEDLINE

Rudner AD and Murray AW. (1996). Curr. Opin. Cell Biol. 8, 773-780. MEDLINE

Statistics and Information Department. (1998). In: Vital Statistics of Japan 1996, Vol. 3. Ministry of Health and Welfare: Tokyo., pp. 384-411.

Takahashi T, Ueda R, Song X, Nishida K, Shinzato M, Namikawa R, Ariyoshi Y, Ota K, Kato K, Nagatsu T, Imaizumi M, Abe T and Takahashi T. (1986). Cancer Res. 46, 4770-4775. MEDLINE

Takahashi T, Nau M, Chiba I, Birrer M, Rosenberg R, Vinocour M, Levitt M, Pass H, Gazdar A and Minna J. (1989). Science 246, 491-494. MEDLINE

Testa JR. (1996). In: Lung Cancer: Principles and Practice. Pass HI, Mitchell JB, Johnson DH and Turrisi AT (eds). Lippincott-Raven: Philadelphia., pp. 55-71.

Weinstein J, Jacobsen FW, Hsu-Chen J, Wu T and Baum LG. (1994). Mol. Cell. Biol. 14, 3350-3363. MEDLINE

Figure 1 Represenative results of flow cytometric analysis of the cell cycle and fluorescence microscopic examination of the mitotic figures of lung cancer cell lines. (a) Flow cytometric analysis of lung cancer cell lines treated with nocodazole for 18 h. Control cell lines with known mitotic checkpoint status are also included. All lines examined show clear cell cycle blocks with DNA content of 4N. (b) Fluorescence microscopic examination of lung cancer cell lines cultured in the presence of nocodazole for 18 h. Accumulation of cells with condensed chromosomes characteristic of a sustained mitotic block is evident in NCI-H460, whereas significantly fewer mitotic cells are present in SK-LU-1 and PC-1

Figure 2 Mitotic indices of lung cancer cell lines in the presence of nocodazole. Control cell lines with known mitotic checkpoint status are also included. Significant reduction of mitotic indices, an indication of an impaired mitotic checkpoint, is evident in four of the nine lung cancer cell lines examined at levels comparable to those in control cell lines with mitotic checkpoint defects

Figure 3 PCR - SSCP analysis of the hsMAD2 gene in lung cancer cell lines. Two types of mobility shifts are present, one corresponding to a silent polymorphism of hsMAD2 (open and solid arrowheads) and the other to the expression of hsMAD2 (arrows)

Figure 4 Sequence analysis of hsMAD2 and RT - PCR analysis of its expression. (a) Nucleotide sequence of hsMAD2 in comparison with hsMAD2. Solid arrows indicate short direct repeats present only in hsMAD2 but absent in hsMAD2. Open arrows indicate PCR primers used to distinguish the hsMAD2 and hsMAD2 transcripts. Solid lines above the hsMAD2 sequence indicate initiation and termination codons of hsMAD2. (b) RT - PCR analysis showing expression of the hsMAD2 transcripts. Omission of reverse transcriptase eliminated RT - PCR amplification of hsMAD2 , whereas RNase-free DNase pretreatment did not